Explore_Keras_3_WIML_2023.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "source": [
        "# Exprole multi backend Keras - Keras 3.0"
      ],
      "metadata": {
        "id": "xCMyWwEhnRFD"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Install dependencies"
      ],
      "metadata": {
        "id": "xDXaQAI6lezX"
      }
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "id": "LMtlfFA9lFMm"
      },
      "outputs": [],
      "source": [
        "! pip install -q keras==3.0"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Set the backend\n",
        "In this tutorial we will use Tensorflow as our backend, you can change it to \"jax\" or \"torch\" for JAX and PyTorch separately."
      ],
      "metadata": {
        "id": "8ojldU2_loSY"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import os\n",
        "\n",
        "# Try picking a different value from \"tensorflow\", \"jax\" and \"torch\".\n",
        "os.environ[\"KERAS_BACKEND\"] = \"tensorflow\""
      ],
      "metadata": {
        "id": "TtgIDp32lmPQ"
      },
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "import tensorflow as tf\n",
        "import keras\n",
        "\n",
        "import numpy as np"
      ],
      "metadata": {
        "id": "VAfXk8pJlxVw"
      },
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Use a Keras layer\n",
        "\n",
        "Layers are callable"
      ],
      "metadata": {
        "id": "kVcH-ogAl4kp"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "layer = keras.layers.Dense(\n",
        "    2,\n",
        "    activation=\"relu\",\n",
        "    bias_initializer=\"zeros\",\n",
        ")\n",
        "data = np.random.uniform(size=[3, 3])\n",
        "layer(data)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "83uHNoDTl7cI",
        "outputId": "98e5dfd8-be30-41d5-b960-0428b18140e5"
      },
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "tensor([[0.0000, 0.0000],\n",
              "        [0.2381, 0.4862],\n",
              "        [0.1233, 0.3835]], grad_fn=<ReluBackward0>)"
            ]
          },
          "metadata": {},
          "execution_count": 4
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Write your custom layer"
      ],
      "metadata": {
        "id": "Gvy8pVx3l_3L"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "class Linear(keras.layers.Layer):\n",
        "    def __init__(self, units):\n",
        "        super().__init__()\n",
        "        self.units = units\n",
        "\n",
        "    def build(self, input_shape):\n",
        "        input_dim = input_shape[-1]\n",
        "        self.w = self.add_weight(\n",
        "            shape=[input_dim, self.units], initializer=\"glorot_uniform\",\n",
        "        )\n",
        "        self.b = self.add_weight(\n",
        "            shape=[self.units,], initializer=\"zeros\"\n",
        "        )\n",
        "\n",
        "    def call(self, inputs):\n",
        "        return keras.ops.matmul(inputs, self.w) + self.b\n",
        "\n",
        "layer = Linear(2)\n",
        "layer(data)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "4ER_ydfkl9tT",
        "outputId": "64139057-e2a8-4b07-c171-0fc3b393b2bf"
      },
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "tensor([[-0.2625,  0.5636],\n",
              "        [-0.9150,  1.2975],\n",
              "        [-0.7664,  1.1620]], grad_fn=<AddBackward0>)"
            ]
          },
          "metadata": {},
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Different ways of building the model\n",
        "\n",
        "*  Sequential API\n",
        "*  [Preferred] Functional API\n",
        "*   Subclassing model\n",
        "\n",
        "In this demo, we are building a 2-class CNN image classifier."
      ],
      "metadata": {
        "id": "SrzyAZWTmE7Z"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Sequential model\n",
        "sequential_model = keras.Sequential([\n",
        "    keras.Input([28, 28, 3]),\n",
        "    keras.layers.Conv2D(8, 2),\n",
        "    keras.layers.MaxPool2D(2),\n",
        "    keras.layers.Flatten(),\n",
        "    keras.layers.Dense(2),\n",
        "    keras.layers.Softmax(),\n",
        "])\n",
        "\n",
        "print(sequential_model.summary())\n",
        "print(sequential_model(np.random.uniform(size=[1, 28, 28, 3])))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 332
        },
        "id": "V0kIpmdymX_V",
        "outputId": "939d3d22-91bf-4b6f-a02e-28dcf3cc1c4d"
      },
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1mModel: \"sequential\"\u001b[0m\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">Model: \"sequential\"</span>\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━┓\n",
              "┃\u001b[1m \u001b[0m\u001b[1mLayer (type)                      \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape                 \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m    Param #\u001b[0m\u001b[1m \u001b[0m┃\n",
              "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━┩\n",
              "│ conv2d (\u001b[38;5;33mConv2D\u001b[0m)                    │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m27\u001b[0m, \u001b[38;5;34m27\u001b[0m, \u001b[38;5;34m8\u001b[0m)             │         \u001b[38;5;34m104\u001b[0m │\n",
              "├────────────────────────────────────┼───────────────────────────────┼─────────────┤\n",
              "│ max_pooling2d (\u001b[38;5;33mMaxPooling2D\u001b[0m)       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m13\u001b[0m, \u001b[38;5;34m13\u001b[0m, \u001b[38;5;34m8\u001b[0m)             │           \u001b[38;5;34m0\u001b[0m │\n",
              "├────────────────────────────────────┼───────────────────────────────┼─────────────┤\n",
              "│ flatten (\u001b[38;5;33mFlatten\u001b[0m)                  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1352\u001b[0m)                  │           \u001b[38;5;34m0\u001b[0m │\n",
              "├────────────────────────────────────┼───────────────────────────────┼─────────────┤\n",
              "│ dense_1 (\u001b[38;5;33mDense\u001b[0m)                    │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2\u001b[0m)                     │       \u001b[38;5;34m2,706\u001b[0m │\n",
              "├────────────────────────────────────┼───────────────────────────────┼─────────────┤\n",
              "│ softmax (\u001b[38;5;33mSoftmax\u001b[0m)                  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2\u001b[0m)                     │           \u001b[38;5;34m0\u001b[0m │\n",
              "└────────────────────────────────────┴───────────────────────────────┴─────────────┘\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━┓\n",
              "┃<span style=\"font-weight: bold\"> Layer (type)                       </span>┃<span style=\"font-weight: bold\"> Output Shape                  </span>┃<span style=\"font-weight: bold\">     Param # </span>┃\n",
              "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━┩\n",
              "│ conv2d (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)                    │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">27</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">27</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">8</span>)             │         <span style=\"color: #00af00; text-decoration-color: #00af00\">104</span> │\n",
              "├────────────────────────────────────┼───────────────────────────────┼─────────────┤\n",
              "│ max_pooling2d (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)       │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">13</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">13</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">8</span>)             │           <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
              "├────────────────────────────────────┼───────────────────────────────┼─────────────┤\n",
              "│ flatten (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Flatten</span>)                  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1352</span>)                  │           <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
              "├────────────────────────────────────┼───────────────────────────────┼─────────────┤\n",
              "│ dense_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                    │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">2</span>)                     │       <span style=\"color: #00af00; text-decoration-color: #00af00\">2,706</span> │\n",
              "├────────────────────────────────────┼───────────────────────────────┼─────────────┤\n",
              "│ softmax (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Softmax</span>)                  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">2</span>)                     │           <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
              "└────────────────────────────────────┴───────────────────────────────┴─────────────┘\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m2,810\u001b[0m (10.98 KB)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Total params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">2,810</span> (10.98 KB)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m2,810\u001b[0m (10.98 KB)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">2,810</span> (10.98 KB)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Non-trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> (0.00 B)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "None\n",
            "tensor([[0.3802, 0.6198]], grad_fn=<SoftmaxBackward0>)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "\n",
        "# Functional model\n",
        "inputs = keras.Input([28, 28, 3])\n",
        "x = keras.layers.Conv2D(8, 2)(inputs)\n",
        "x = keras.layers.MaxPool2D(2)(x)\n",
        "x = keras.layers.Flatten()(x)\n",
        "x = keras.layers.Dense(2)(x)\n",
        "outputs = keras.layers.Softmax()(x)\n",
        "\n",
        "functional_model = keras.Model(inputs=inputs, outputs=outputs)\n",
        "print(functional_model.summary())\n",
        "print(functional_model(np.random.uniform(size=[1, 28, 28, 3])))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 365
        },
        "id": "1-Kx6vBAmbbs",
        "outputId": "623e62ed-6ac6-4c47-a844-db64d13593ab"
      },
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1mModel: \"functional_2\"\u001b[0m\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">Model: \"functional_2\"</span>\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━┓\n",
              "┃\u001b[1m \u001b[0m\u001b[1mLayer (type)                      \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape                 \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m    Param #\u001b[0m\u001b[1m \u001b[0m┃\n",
              "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━┩\n",
              "│ input_layer_1 (\u001b[38;5;33mInputLayer\u001b[0m)         │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m3\u001b[0m)             │           \u001b[38;5;34m0\u001b[0m │\n",
              "├────────────────────────────────────┼───────────────────────────────┼─────────────┤\n",
              "│ conv2d_1 (\u001b[38;5;33mConv2D\u001b[0m)                  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m27\u001b[0m, \u001b[38;5;34m27\u001b[0m, \u001b[38;5;34m8\u001b[0m)             │         \u001b[38;5;34m104\u001b[0m │\n",
              "├────────────────────────────────────┼───────────────────────────────┼─────────────┤\n",
              "│ max_pooling2d_1 (\u001b[38;5;33mMaxPooling2D\u001b[0m)     │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m13\u001b[0m, \u001b[38;5;34m13\u001b[0m, \u001b[38;5;34m8\u001b[0m)             │           \u001b[38;5;34m0\u001b[0m │\n",
              "├────────────────────────────────────┼───────────────────────────────┼─────────────┤\n",
              "│ flatten_1 (\u001b[38;5;33mFlatten\u001b[0m)                │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1352\u001b[0m)                  │           \u001b[38;5;34m0\u001b[0m │\n",
              "├────────────────────────────────────┼───────────────────────────────┼─────────────┤\n",
              "│ dense_2 (\u001b[38;5;33mDense\u001b[0m)                    │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2\u001b[0m)                     │       \u001b[38;5;34m2,706\u001b[0m │\n",
              "├────────────────────────────────────┼───────────────────────────────┼─────────────┤\n",
              "│ softmax_1 (\u001b[38;5;33mSoftmax\u001b[0m)                │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2\u001b[0m)                     │           \u001b[38;5;34m0\u001b[0m │\n",
              "└────────────────────────────────────┴───────────────────────────────┴─────────────┘\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━┓\n",
              "┃<span style=\"font-weight: bold\"> Layer (type)                       </span>┃<span style=\"font-weight: bold\"> Output Shape                  </span>┃<span style=\"font-weight: bold\">     Param # </span>┃\n",
              "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━┩\n",
              "│ input_layer_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">InputLayer</span>)         │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">28</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">28</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)             │           <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
              "├────────────────────────────────────┼───────────────────────────────┼─────────────┤\n",
              "│ conv2d_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)                  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">27</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">27</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">8</span>)             │         <span style=\"color: #00af00; text-decoration-color: #00af00\">104</span> │\n",
              "├────────────────────────────────────┼───────────────────────────────┼─────────────┤\n",
              "│ max_pooling2d_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)     │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">13</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">13</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">8</span>)             │           <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
              "├────────────────────────────────────┼───────────────────────────────┼─────────────┤\n",
              "│ flatten_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Flatten</span>)                │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1352</span>)                  │           <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
              "├────────────────────────────────────┼───────────────────────────────┼─────────────┤\n",
              "│ dense_2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                    │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">2</span>)                     │       <span style=\"color: #00af00; text-decoration-color: #00af00\">2,706</span> │\n",
              "├────────────────────────────────────┼───────────────────────────────┼─────────────┤\n",
              "│ softmax_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Softmax</span>)                │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">2</span>)                     │           <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
              "└────────────────────────────────────┴───────────────────────────────┴─────────────┘\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m2,810\u001b[0m (10.98 KB)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Total params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">2,810</span> (10.98 KB)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m2,810\u001b[0m (10.98 KB)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">2,810</span> (10.98 KB)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Non-trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> (0.00 B)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "None\n",
            "tensor([[0.5268, 0.4732]], grad_fn=<SoftmaxBackward0>)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Multi-input model.\n",
        "input_1 = keras.Input([28, 28, 3])\n",
        "input_2 = keras.Input([28, 28, 3])\n",
        "x_1 = keras.layers.Conv2D(8, 2)(input_1)\n",
        "x_2 = keras.layers.Conv2D(8, 2)(input_2)\n",
        "x = keras.layers.MaxPool2D(2)(x_1 + x_2)\n",
        "x = keras.layers.Flatten()(x)\n",
        "x = keras.layers.Dense(2)(x)\n",
        "outputs = keras.layers.Softmax()(x)\n",
        "\n",
        "multi_input_functional_model = keras.Model(inputs=inputs, outputs=outputs)\n",
        "print(multi_input_functional_model.summary())"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 562
        },
        "id": "9sJGmu_RmfWs",
        "outputId": "f7c1af6a-628b-4e19-94ce-8e019aa22b52"
      },
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1mModel: \"functional_4\"\u001b[0m\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">Model: \"functional_4\"</span>\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━┓\n",
              "┃\u001b[1m \u001b[0m\u001b[1mLayer (type)             \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape          \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m   Param #\u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mConnected to              \u001b[0m\u001b[1m \u001b[0m┃\n",
              "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n",
              "│ input_layer_2             │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m3\u001b[0m)      │          \u001b[38;5;34m0\u001b[0m │ -                          │\n",
              "│ (\u001b[38;5;33mInputLayer\u001b[0m)              │                        │            │                            │\n",
              "├───────────────────────────┼────────────────────────┼────────────┼────────────────────────────┤\n",
              "│ input_layer_3             │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m3\u001b[0m)      │          \u001b[38;5;34m0\u001b[0m │ -                          │\n",
              "│ (\u001b[38;5;33mInputLayer\u001b[0m)              │                        │            │                            │\n",
              "├───────────────────────────┼────────────────────────┼────────────┼────────────────────────────┤\n",
              "│ conv2d_2 (\u001b[38;5;33mConv2D\u001b[0m)         │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m27\u001b[0m, \u001b[38;5;34m27\u001b[0m, \u001b[38;5;34m8\u001b[0m)      │        \u001b[38;5;34m104\u001b[0m │ input_layer_2[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]        │\n",
              "├───────────────────────────┼────────────────────────┼────────────┼────────────────────────────┤\n",
              "│ conv2d_3 (\u001b[38;5;33mConv2D\u001b[0m)         │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m27\u001b[0m, \u001b[38;5;34m27\u001b[0m, \u001b[38;5;34m8\u001b[0m)      │        \u001b[38;5;34m104\u001b[0m │ input_layer_3[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]        │\n",
              "├───────────────────────────┼────────────────────────┼────────────┼────────────────────────────┤\n",
              "│ add (\u001b[38;5;33mAdd\u001b[0m)                 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m27\u001b[0m, \u001b[38;5;34m27\u001b[0m, \u001b[38;5;34m8\u001b[0m)      │          \u001b[38;5;34m0\u001b[0m │ conv2d_2[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m],            │\n",
              "│                           │                        │            │ conv2d_3[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]             │\n",
              "├───────────────────────────┼────────────────────────┼────────────┼────────────────────────────┤\n",
              "│ max_pooling2d_2           │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m13\u001b[0m, \u001b[38;5;34m13\u001b[0m, \u001b[38;5;34m8\u001b[0m)      │          \u001b[38;5;34m0\u001b[0m │ add[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]                  │\n",
              "│ (\u001b[38;5;33mMaxPooling2D\u001b[0m)            │                        │            │                            │\n",
              "├───────────────────────────┼────────────────────────┼────────────┼────────────────────────────┤\n",
              "│ flatten_2 (\u001b[38;5;33mFlatten\u001b[0m)       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1352\u001b[0m)           │          \u001b[38;5;34m0\u001b[0m │ max_pooling2d_2[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]      │\n",
              "├───────────────────────────┼────────────────────────┼────────────┼────────────────────────────┤\n",
              "│ dense_3 (\u001b[38;5;33mDense\u001b[0m)           │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2\u001b[0m)              │      \u001b[38;5;34m2,706\u001b[0m │ flatten_2[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]            │\n",
              "├───────────────────────────┼────────────────────────┼────────────┼────────────────────────────┤\n",
              "│ input_layer_1             │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m3\u001b[0m)      │          \u001b[38;5;34m0\u001b[0m │ -                          │\n",
              "│ (\u001b[38;5;33mInputLayer\u001b[0m)              │                        │            │                            │\n",
              "├───────────────────────────┼────────────────────────┼────────────┼────────────────────────────┤\n",
              "│ softmax_2 (\u001b[38;5;33mSoftmax\u001b[0m)       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2\u001b[0m)              │          \u001b[38;5;34m0\u001b[0m │ dense_3[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]              │\n",
              "└───────────────────────────┴────────────────────────┴────────────┴────────────────────────────┘\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━┓\n",
              "┃<span style=\"font-weight: bold\"> Layer (type)              </span>┃<span style=\"font-weight: bold\"> Output Shape           </span>┃<span style=\"font-weight: bold\">    Param # </span>┃<span style=\"font-weight: bold\"> Connected to               </span>┃\n",
              "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n",
              "│ input_layer_2             │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">28</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">28</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)      │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ -                          │\n",
              "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">InputLayer</span>)              │                        │            │                            │\n",
              "├───────────────────────────┼────────────────────────┼────────────┼────────────────────────────┤\n",
              "│ input_layer_3             │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">28</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">28</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)      │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ -                          │\n",
              "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">InputLayer</span>)              │                        │            │                            │\n",
              "├───────────────────────────┼────────────────────────┼────────────┼────────────────────────────┤\n",
              "│ conv2d_2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)         │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">27</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">27</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">8</span>)      │        <span style=\"color: #00af00; text-decoration-color: #00af00\">104</span> │ input_layer_2[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]        │\n",
              "├───────────────────────────┼────────────────────────┼────────────┼────────────────────────────┤\n",
              "│ conv2d_3 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)         │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">27</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">27</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">8</span>)      │        <span style=\"color: #00af00; text-decoration-color: #00af00\">104</span> │ input_layer_3[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]        │\n",
              "├───────────────────────────┼────────────────────────┼────────────┼────────────────────────────┤\n",
              "│ add (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Add</span>)                 │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">27</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">27</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">8</span>)      │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ conv2d_2[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>],            │\n",
              "│                           │                        │            │ conv2d_3[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]             │\n",
              "├───────────────────────────┼────────────────────────┼────────────┼────────────────────────────┤\n",
              "│ max_pooling2d_2           │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">13</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">13</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">8</span>)      │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ add[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]                  │\n",
              "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)            │                        │            │                            │\n",
              "├───────────────────────────┼────────────────────────┼────────────┼────────────────────────────┤\n",
              "│ flatten_2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Flatten</span>)       │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1352</span>)           │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ max_pooling2d_2[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]      │\n",
              "├───────────────────────────┼────────────────────────┼────────────┼────────────────────────────┤\n",
              "│ dense_3 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)           │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">2</span>)              │      <span style=\"color: #00af00; text-decoration-color: #00af00\">2,706</span> │ flatten_2[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]            │\n",
              "├───────────────────────────┼────────────────────────┼────────────┼────────────────────────────┤\n",
              "│ input_layer_1             │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">28</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">28</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)      │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ -                          │\n",
              "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">InputLayer</span>)              │                        │            │                            │\n",
              "├───────────────────────────┼────────────────────────┼────────────┼────────────────────────────┤\n",
              "│ softmax_2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Softmax</span>)       │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">2</span>)              │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ dense_3[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]              │\n",
              "└───────────────────────────┴────────────────────────┴────────────┴────────────────────────────┘\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m2,914\u001b[0m (11.38 KB)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Total params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">2,914</span> (11.38 KB)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m2,914\u001b[0m (11.38 KB)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">2,914</span> (11.38 KB)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Non-trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> (0.00 B)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "None\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Subclassing `keras.Model`\n",
        "\n",
        "class MyModel(keras.Model):\n",
        "    def __init__(self, *args, **kwargs):\n",
        "        super().__init__(*args, **kwargs)\n",
        "        self._conv = keras.layers.Conv2D(8, 2)\n",
        "        self._max_pool = keras.layers.MaxPool2D(2)\n",
        "        self._flatten = keras.layers.Flatten()\n",
        "        self._dense = keras.layers.Dense(2)\n",
        "        self._softmax = keras.layers.Softmax()\n",
        "\n",
        "    def call(self, inputs):\n",
        "        x = self._conv(inputs)\n",
        "        x = self._max_pool(x)\n",
        "        x = self._flatten(x)\n",
        "        x = self._dense(x)\n",
        "        return self._softmax(x)\n",
        "\n",
        "subclass_model = MyModel()\n",
        "\n",
        "print(subclass_model(np.random.uniform(size=[1, 28, 28, 3])))\n",
        "print(subclass_model.summary())\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 332
        },
        "id": "SspFkvbbmk7w",
        "outputId": "21c78a7b-b011-4067-a78a-28e04429bc13"
      },
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "tensor([[0.5375, 0.4625]], grad_fn=<SoftmaxBackward0>)\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1mModel: \"my_model\"\u001b[0m\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">Model: \"my_model\"</span>\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━┓\n",
              "┃\u001b[1m \u001b[0m\u001b[1mLayer (type)                      \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape                 \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m    Param #\u001b[0m\u001b[1m \u001b[0m┃\n",
              "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━┩\n",
              "│ conv2d_4 (\u001b[38;5;33mConv2D\u001b[0m)                  │ ?                             │         \u001b[38;5;34m104\u001b[0m │\n",
              "├────────────────────────────────────┼───────────────────────────────┼─────────────┤\n",
              "│ max_pooling2d_3 (\u001b[38;5;33mMaxPooling2D\u001b[0m)     │ ?                             │           \u001b[38;5;34m0\u001b[0m │\n",
              "├────────────────────────────────────┼───────────────────────────────┼─────────────┤\n",
              "│ flatten_3 (\u001b[38;5;33mFlatten\u001b[0m)                │ ?                             │           \u001b[38;5;34m0\u001b[0m │\n",
              "├────────────────────────────────────┼───────────────────────────────┼─────────────┤\n",
              "│ dense_4 (\u001b[38;5;33mDense\u001b[0m)                    │ ?                             │       \u001b[38;5;34m2,706\u001b[0m │\n",
              "├────────────────────────────────────┼───────────────────────────────┼─────────────┤\n",
              "│ softmax_3 (\u001b[38;5;33mSoftmax\u001b[0m)                │ ?                             │           \u001b[38;5;34m0\u001b[0m │\n",
              "└────────────────────────────────────┴───────────────────────────────┴─────────────┘\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━┓\n",
              "┃<span style=\"font-weight: bold\"> Layer (type)                       </span>┃<span style=\"font-weight: bold\"> Output Shape                  </span>┃<span style=\"font-weight: bold\">     Param # </span>┃\n",
              "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━┩\n",
              "│ conv2d_4 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)                  │ ?                             │         <span style=\"color: #00af00; text-decoration-color: #00af00\">104</span> │\n",
              "├────────────────────────────────────┼───────────────────────────────┼─────────────┤\n",
              "│ max_pooling2d_3 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)     │ ?                             │           <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
              "├────────────────────────────────────┼───────────────────────────────┼─────────────┤\n",
              "│ flatten_3 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Flatten</span>)                │ ?                             │           <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
              "├────────────────────────────────────┼───────────────────────────────┼─────────────┤\n",
              "│ dense_4 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                    │ ?                             │       <span style=\"color: #00af00; text-decoration-color: #00af00\">2,706</span> │\n",
              "├────────────────────────────────────┼───────────────────────────────┼─────────────┤\n",
              "│ softmax_3 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Softmax</span>)                │ ?                             │           <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
              "└────────────────────────────────────┴───────────────────────────────┴─────────────┘\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m2,810\u001b[0m (10.98 KB)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Total params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">2,810</span> (10.98 KB)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m2,810\u001b[0m (10.98 KB)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">2,810</span> (10.98 KB)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Non-trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> (0.00 B)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "None\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Models are callable.\n",
        "`keras.Model` instances are also callable. You can call models as if it is a function, no matter which way you used to build the model."
      ],
      "metadata": {
        "id": "3k4idnXtmoxd"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "sample_data = np.random.uniform(size=[1, 28, 28, 3])\n",
        "\n",
        "print(sequential_model(sample_data))\n",
        "print(functional_model(sample_data))\n",
        "print(subclass_model(sample_data))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "OAl6KDgfmmSC",
        "outputId": "6aded2ac-d103-4df9-c3e8-778532383871"
      },
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "tensor([[0.3852, 0.6148]], grad_fn=<SoftmaxBackward0>)\n",
            "tensor([[0.5532, 0.4468]], grad_fn=<SoftmaxBackward0>)\n",
            "tensor([[0.5829, 0.4171]], grad_fn=<SoftmaxBackward0>)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "sliced_model = keras.Model(\n",
        "    inputs=functional_model.inputs,\n",
        "    outputs=functional_model.get_layer(\"conv2d_1\").output,\n",
        ")\n",
        "sliced_model.summary()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 199
        },
        "id": "Hy0c04GtmwqZ",
        "outputId": "02517ae4-0749-48df-9216-a5f6636db66a"
      },
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1mModel: \"functional_6\"\u001b[0m\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">Model: \"functional_6\"</span>\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━┓\n",
              "┃\u001b[1m \u001b[0m\u001b[1mLayer (type)                      \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape                 \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m    Param #\u001b[0m\u001b[1m \u001b[0m┃\n",
              "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━┩\n",
              "│ input_layer_1 (\u001b[38;5;33mInputLayer\u001b[0m)         │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m3\u001b[0m)             │           \u001b[38;5;34m0\u001b[0m │\n",
              "├────────────────────────────────────┼───────────────────────────────┼─────────────┤\n",
              "│ conv2d_1 (\u001b[38;5;33mConv2D\u001b[0m)                  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m27\u001b[0m, \u001b[38;5;34m27\u001b[0m, \u001b[38;5;34m8\u001b[0m)             │         \u001b[38;5;34m104\u001b[0m │\n",
              "└────────────────────────────────────┴───────────────────────────────┴─────────────┘\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━┓\n",
              "┃<span style=\"font-weight: bold\"> Layer (type)                       </span>┃<span style=\"font-weight: bold\"> Output Shape                  </span>┃<span style=\"font-weight: bold\">     Param # </span>┃\n",
              "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━┩\n",
              "│ input_layer_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">InputLayer</span>)         │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">28</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">28</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)             │           <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
              "├────────────────────────────────────┼───────────────────────────────┼─────────────┤\n",
              "│ conv2d_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)                  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">27</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">27</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">8</span>)             │         <span style=\"color: #00af00; text-decoration-color: #00af00\">104</span> │\n",
              "└────────────────────────────────────┴───────────────────────────────┴─────────────┘\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m104\u001b[0m (416.00 B)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Total params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">104</span> (416.00 B)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m104\u001b[0m (416.00 B)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">104</span> (416.00 B)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Non-trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> (0.00 B)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Train the model\n",
        "Now we have defined the model. For actual training, we need the following pieces:\n",
        "\n",
        "* Data\n",
        "* Loss/Target function\n",
        "* Optimizer"
      ],
      "metadata": {
        "id": "0cTr-gLHm4xj"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "\n",
        "# Prepare data for a 2-class classification.\n",
        "data = np.random.uniform(size=[20, 28, 28, 3])\n",
        "label = np.random.randint(2, size=20)"
      ],
      "metadata": {
        "id": "u1QKXbgum-Xr"
      },
      "execution_count": 12,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "\n",
        "# Let's use functional model defined above.\n",
        "model = functional_model"
      ],
      "metadata": {
        "id": "VB9EJ6ZDm_cf"
      },
      "execution_count": 13,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "model.compile(\n",
        "    loss=keras.losses.SparseCategoricalCrossentropy(),\n",
        "    optimizer=keras.optimizers.Adam(0.001),\n",
        "    metrics=[keras.metrics.SparseCategoricalAccuracy()],\n",
        ")"
      ],
      "metadata": {
        "id": "nVUkiDmunCyu"
      },
      "execution_count": 14,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "model.fit(\n",
        "    data,\n",
        "    label,\n",
        "    batch_size=5,\n",
        "    epochs=2,\n",
        ")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "24Cy-bzQnFNS",
        "outputId": "5740f119-9e02-4aeb-ba94-02151a21e007"
      },
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/2\n",
            "\u001b[1m4/4\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 16ms/step - loss: 0.7391 - sparse_categorical_accuracy: 0.3800 \n",
            "Epoch 2/2\n",
            "\u001b[1m4/4\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 23ms/step - loss: 0.6861 - sparse_categorical_accuracy: 0.5333\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<keras.src.callbacks.history.History at 0x7b6d61bb0f70>"
            ]
          },
          "metadata": {},
          "execution_count": 15
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Save your model"
      ],
      "metadata": {
        "id": "3Dlbgnvun4jo"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "\n",
        "# Save the weights.\n",
        "model.save_weights(\"dummy.weights.h5\")"
      ],
      "metadata": {
        "id": "H-BVKUIun5s3"
      },
      "execution_count": 23,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "sequential_model.load_weights(\"dummy.weights.h5\")"
      ],
      "metadata": {
        "id": "YVxoWKzhn7nR"
      },
      "execution_count": 24,
      "outputs": []
    }
  ]
}